Dr. Howes opened up his lovely home to host journal club this month on October 5, 2016 – it was an evening of pizza, ice cream, and enlightening discussion. As usual, two articles were featured. The commentary below, written by Kristen, was staff reviewed by Dan.
Endovascular thrombectomy after large vessel ischaemic stroke: a meta-analysis of individual patient data from five randomized trails
by Goyal et al. published in the Lancet 2016.
Dr. Keegan Selby presented the resident chosen article listed above, providing an excellent summary of the “biggest thing to ever happen to neurology” in the context of current practices. To quote Dr. Rob Brison directly, the authors used a “really cool” technique to combine raw patient data from five previous studies, MR CLEAN, ESCAPE, SWIFT PRIME, REVASCAT, and EXTEND IA trials. It is amazing to see the possibilities with research collaboration such as this! We were lucky to have a strong staff presence to help us understand the basics of the mixed methods modelling used. The study is not a traditional meta-analysis and avoids much of the bias because patient level data was used and combined, rather than data that had already been sorted through and grouped. Importantly, authors did account for between-trial variation and were able to provide a more powerful and reliable conclusion than each individual study alone with these statistical techniques. Interestingly, the National Institute of Research is hoping to make all raw data available online from authors who publish under their grants in the future. Just think of the possibilities!
In the end, we agree with the authors’ conclusion – EVT seems like a great idea to reduce disability in patients with large vessel anterior circulation ischemic strokes if you live close enough to a center proficient in this technique in a system that can afford it. Here in Kingston, we are capable – what remains to be determined is whether this is something that is feasible and ethical for KGH.
[Extra tidbit: 5 patients have undergone EVT at KGH – it is currently available during business hours to the optimal candidates. Bottom line: discuss with the stroke team if you encounter a patient with a large anterior ischemic stroke in the ED.]
Intensive Blood-Pressure Lowering in Patients with Acute Cerebral Hemorrhage (ATACH-2)
by Qureshi et al. published in NEJM 2016
Dr. Howes led the dialogue through the chosen staff article, ATACH-2 . The authors conducted a randomized, multicenter, open-label trial to examine the effects of intensive blood pressure control in acute cerebral hemorrhage, comparing a target of 110-140mmHg to 140-179mmHg using IV nicardipine – and stopped enrollment early due to futility after interim analysis. I won’t go into all of the details but we had rich discussion surrounding the standard treatment of acute cerebral hemorrhage at KGH, the generalizability of the data to our population, the generous enrollment criteria, the inadequate power of the study (keeping in mind it was stopped early), and the practical difficulty and reader uncertainty surrounding specifics of blood pressure control in this study. In the end, as the authors conclude, this will not change our current practice in the management of acute cerebral hemorrhage. A few take home points were emphasized that can be applied to any article:
- Always look at the estimate of treatment effect – the authors used a very optimistic 10% difference in likelihood of death or disability at 3 months between their intervention and standard treatment to calculate the power needed at the outset of this study.
- A superiority trial ≠ an equivalency trial ≠ an inferiority trial – refer to a previous post by Eve, Arrests and Ankles at Astors, for a refresher
- Be alert to misleading conclusions – Dan used a personal anecdote of a special stuffed animal to remind us to avoid the “blue dog” false conclusion in our research and analyses; just because you don’t find blue dog in your search, does not mean that you can conclude for certain that blue dog is not in the house.